Randomised comparison of leucocyte-depleted versus buffy-coat-poor blood transfusion and complications after colorectal surgery

Background Allogeneic blood transfusion is associated with an increased frequency of postoperative infection. We studied whether such events can be avoided by the intraoperative and postoperative use of leucocyte-depleted blood. Methods 589 consecutive patients scheduled for elective colorectal surgery were randomised to receive buffy-coat poor (n=299) or filtered leucocyte-depleted red-cells (n=290) when transfusion was indicated. 260 patients actually received blood transfusion. Three patients were excluded from analysis. Findings The 142 patients randomised to and transfused with buffy-coat-poor blood had a significantly higher frequency of wound infections and intra-abdominal abscesses than the 155 patients who were allocated to this group but who were not transfused. (12 vs 1%, p<0.0001) and (5 vs 0%, p=0.005), respectively. Those receiving buffy-coat-poor blood also had a significantly higher frequency of wound infections and intra-abdominal abscesses than the 118 randomised to and receiving leucocyte-depleted blood (12 vs 0%, p<0.0001) and (5 vs 0%, p=0.017), respectively. Postoperative pneumonia occurred at a significantly higher rate in patients receiving buffy-coat-poor blood: 23 versus 3% in non-transfused patients (p<0.0001), and 23 versus 3% in patients transfused with leucocyte-depleted blood (p<0.001). The mortality rate between the groups was not statistically different. Interpretation Our data suggests that the association between allogeneic blood transfusion and postoperative infection is limited to allogeneic blood products that are not adequately depleted of immunosuppressive leucocytes. This undesirable effect can be reduced by leucocyte depletion with high-efficiency filters.